Database sharding strategies in HLD - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is database sharding?

Database sharding is a technique to split a large database into smaller, faster, and more manageable parts called shards. Each shard holds a portion of the data, helping improve performance and scalability.

beginner

Explain horizontal sharding.

Horizontal sharding divides data by rows. Each shard contains a subset of rows from the same table, often based on a key like user ID. This spreads the load across multiple servers.

intermediate

What is vertical sharding and when is it used?

Vertical sharding splits data by columns or tables. Different shards hold different parts of the schema. It is useful when some tables or columns are accessed more frequently or require different resources.

intermediate

Describe directory-based sharding.

Directory-based sharding uses a lookup service to map each data item to its shard. This allows flexible shard assignment but adds a dependency on the directory for every query.

advanced

What are common challenges of database sharding?

Challenges include data rebalancing when shards grow unevenly, handling cross-shard queries, maintaining consistency, and increased complexity in application logic.

Which sharding strategy splits data by rows across multiple shards?

A Directory-based sharding

B Vertical sharding

C Horizontal sharding

D Functional sharding

Horizontal sharding divides data by rows, distributing subsets of rows across shards.

What is a key disadvantage of directory-based sharding?

A It duplicates data across shards

B It cannot handle large datasets

C It only works with vertical sharding

D It requires a lookup service for every query

Directory-based sharding depends on a lookup service to find the shard, adding overhead and a potential single point of failure.

Vertical sharding splits data based on:

A Columns or tables

B User IDs

C Rows

D Random distribution

Vertical sharding divides data by columns or tables, separating parts of the schema into different shards.

Which problem is NOT typically caused by sharding?

A Cross-shard query complexity

B Automatic data backup

C Increased application complexity

D Data rebalancing

Automatic data backup is not a direct problem caused by sharding; it is usually handled separately.

What is the main goal of database sharding?

A To improve performance and scalability by splitting data

B To reduce database size by deleting data

C To encrypt data for security

D To merge multiple databases into one

Sharding splits data into smaller parts to improve performance and allow the system to scale.

Describe the main types of database sharding strategies and when to use each.

Think about how data is divided: by rows, columns, or using a directory.

What challenges arise from implementing database sharding and how can they affect system design?

Consider operational and development difficulties after splitting data.